Predictive modeling of selected trace elements in groundwater using hybrid algorithms of iterative classifier optimizer

A multi-step approach to evaluate the sustainable use of groundwater resources for human consumption and agriculture

The rapid decline in both quality and availability of freshwater resources on our planet necessitates their thorough assessment to ensure sustainable usage. The growing demand for water in industrial, agricultural, and domestic sectors poses significant challenges to managing both surface and groundwater resources. This study tests and proposes a hybrid evaluation approach to determine Groundwater Quality Indices (GQIs) for irrigation (IRRI), seawater intrusion (SWI), and potability (POT), finalized to the spatial distribution of groundwater suitability involving water quality indicator along with hydrogeological and socio-economic factors. Mean Decrease Accuracy (MDA) and Information Gain Ratio (IGR) were used to state the importance of chosen factors such as level of groundwater above the sea, thickness of the aquifer, land cover, distance from coastline, silt soil content, recharge, distance from river and lagoons, depth to water table from ground, distance from agricultural wells, hydraulic conductivity, and lithology for each quality index, separately. The results of both methods showed that recharge is the most important parameter for GQIIRRI and GQIPOT, while the distance from the coastline and the rivers, are the most important for GQISWI. The spatial modelling of GQIIRRI and GQIPOT in the study area has been achieved applying three machine learning (ML) algorithms: the Boosted Regression Tree (BRT), the Random Forest (RF), and the Support Vector Machine (SVM). Validation results showed that RF has the highest prediction for GQIIRRI, while the SVM model has the highest prediction for the GQIPOT index. It is worth to mention that the future utilization and testing of new algorithms could produce even better results. Finally, GQIIRRI and GQIPOT were combined and compared using two combine and overlay methods to prepare a hybrid map of multi-GQIs. The results showed that 69% of the study area is suitable for irrigation and potable use, due to both geogenic and anthropogenic activities which contribute to make some water resources unsuitable for either use. Specifically, the northern, western, and eastern portions of the study area are in the "very high and high quality" classes while the southern portion shows "very low and low quality" classes. In conclusion, the developed map and approach can serve as a practical guide for enhancing groundwater management, identifying suitable areas for various uses and pinpointing regions requiring improved management practices.

Coupling SWAT and DPSIR models for groundwater management in Mediterranean catchments

Groundwater is an essential natural resource and has a significant role in human and environmental health as well as in the economy. Management of subsurface storage remains an important option to meet the combined demands of humans and ecosystems. The increasing need to find multi-purpose solutions to address water scarcity is a global challenge. Thus, the interactions leading to surface runoff and groundwater recharge have received particular attention over the last decades. Additionally, new methods are developed to incorporate the spatial-temporal variation of recharge in groundwater modeling. In this study, groundwater recharge was spatiotemporally quantified using the Soil and Water Assessment Tool (SWAT) in the Upper Volturno-Calore hydrological basin in Italy and the results were compared with other two basins in Greece (Anthemountas and Mouriki). SWAT model was applied in actual and future projections (2022-2040) using the Representative Concentration Pathway (RCP) 4.5 emissions scenario to evaluate changes in precipitation and assess the future hydrologic conditions, along with, the Driving Force-Pressure-State-Impact-Response (DPSIR) framework that was applied in all the basins as a low-cost analysis of integrated physical, social, natural, and economic factors. According to the results, no significant variations in runoff are predicted in the Upper Volturno-Calore basin for the period 2020-2040 while the potential evapotranspiration percentage varies from 50.1% to 74.3% and infiltration around 5%. The limited primary data constitutes the main pressure in all sites and exaggerates the uncertainty of future projections.

Hair and cord blood element levels and their relationship with air pollution, dietary intake, gestational diabetes mellitus, and infant neurodevelopment

BACKGROUND & AIMS

Exposure to a range of elements, air pollution, and specific dietary components in pregnancy has variously been associated with gestational diabetes mellitus (GDM) risk or infant neurodevelopmental problems. We measured a range of pregnancy exposures in maternal hair and/or infant cord serum and tested their relationship to GDM and infant neurodevelopment.

METHODS

A total of 843 pregnant women (GDM = 224, Non-GDM = 619) were selected from the Complex Lipids in Mothers and Babies cohort study. Forty-eight elements in hair and cord serum were quantified using inductively coupled plasma-mass spectrometry analysis. Binary logistic regression was used to estimate the associations between hair element concentrations and GDM risk, while multiple linear regression was performed to analyze the relationship between hair/cord serum elements and air pollutants, diet exposures, and Bayley Scales of infant neurodevelopment at 12 months of age.

RESULTS

After adjusting for maternal age, BMI, and primiparity, we observed that fourteen elements in maternal hair were associated with a significantly increased risk of GDM, particularly Ta (OR = 9.49, 95% CI: 6.71, 13.42), Re (OR = 5.21, 95% CI: 3.84, 7.07), and Se (OR = 5.37, 95% CI: 3.48, 8.28). In the adjusted linear regression model, three elements (Rb, Er, and Tm) in maternal hair and infant cord serum were negatively associated with Mental Development Index scores. For dietary exposures, elements were positively associated with noodles (Nb), sweetened beverages (Rb), poultry (Cs), oils and condiments (Ca), and other seafood (Gd). In addition, air pollutants PM2.5 (LUR) and PM10 were negatively associated with Ta and Re in maternal hair.

CONCLUSIONS

Our findings highlight the potential influence of maternal element exposure on GDM risk and infant neurodevelopment. We identified links between levels of these elements in both maternal hair and infant cord serum related to air pollutants and dietary factors.

Effects of elevated arsenic and nitrate concentrations on groundwater resources in deltaic region of Sundarban Ramsar site, Indo-Bangladesh region

An attempt has been adopted to predict the As and NO₃^- concentration in groundwater (GW) in fast-growing coastal Ramsar region in eastern India. This study is focused to evaluate the As and NO₃^- vulnerable areas of coastal belts of the Indo-Bangladesh Ramsar site a hydro-geostrategic region of the world by using advanced ensemble ML techniques including NB-RF, NB-SVM and NB-Bagging. A total of 199 samples were collected from the entire study area for utilizing the 12 GWQ conditioning factors. The predicted results are certified that NB-Bagging the most suitable and preferable model in this current research. The vulnerability of As and NO₃^- concentration shows that most of the areas are highly vulnerable to As and low to moderately vulnerable to NO₃. The reliable findings of this present study will help the management authorities and policymakers in taking preventive measures in reducing the vulnerability of water resources and corresponding health risks.

A novel hybrid random forest linear model approach for forecasting groundwater fluoride contamination

Groundwater quality in the Datong basin is threatened by high fluoride contamination. Laboratory analysis is a standard method for estimating groundwater quality parameters, which is expensive and time-consuming. Therefore, this paper proposes a hybrid random forest linear model (HRFLM) as a novel approach for estimating groundwater fluoride contamination. Light gradient boosting (LightGBM), random forest (RF), and extreme gradient boosting (Xgboost) were also employed in comparison with HRFLM for predicting fluoride contamination in groundwater. 202 groundwater samples were collected to draw up the performance capability of several models in forecasting subsurface water fluoride contamination. The performance of the models was assessed utilizing the receiver operating characteristic (ROC) area under the curve (AUC) and the confusion matrix (CM). The CM results reveal that with nine predictor variables, the hybrid HRFLM achieved an accuracy of 95%, outperforming the Xgboost, LightGBM, and RF models, which attained 88%, 88%, and 85%, respectively. Likewise, the AUC results of the hybrid HRFLM show high performance with an AUC of 0.98 compared to Xgboost, LightGBM, and RF, which achieved an AUC of 0.95, 0.90, and 0.88, respectively. The study demonstrates that the HRFLM can be applied as an advanced approach for groundwater fluoride contamination prediction in the Datong basin and could be adopted in various areas facing a similar challenge.

A partial least squares-path model of environmental degradation in the Paraopeba River, for rainy seasons after the rupture of B1 tailings dam, Brumadinho, Brazil

The present study aimed to investigate the rupture of B1 tailings dam of Córrego do Feijão mine, which drastically affected the region of Brumadinho (Minas Gerais, Brazil). The contamination of water resources reached 155.3 km from the dam site. In the river channel, high concentrations of Mn, Al, As and Fe were detected and correlated to the spillage of the tailings in the river. The presence of the tailings also affected the chlorophyll-a content in the water, as well as the reflectance of riparian forests. With the increase of metal(oid) concentrations above permitted levels, water management authorities suspended the use of Paraopeba River as resource in the impacted areas, namely the drinking water supply to the Metropolitan region of Belo Horizonte. This study aimed to evaluate possible links between tailings distribution, river water quality, and environmental degradation, which worked as latent variables in partial least squares regression models. The latent variables were represented by numerous physical and chemical parameters of water and sediment, measured four times in 22 locations during the rainy season of 2019, in addition to stream flow and to NDVI evaluated in satellite images processed daily. The modeling results suggested a relationship between river flow turbulence and increased arsenic release from sand fractions, as well as desorption of Mn from metal oxides, both representing causes of water quality reduction. They also revealed increasing iron concentrations affecting the forest NDVI (greening), which was interpreted as environmental degradation. The increase of chlorophyll-a concentrations (related with turbidity decreases), as well as the increase of river flows (responsible for dilution effects), seemed to work out as attenuators of degradation. Although applied to a specific site, our modeling approach can be transposed to equivalent dam failures and climate contexts, helping water resource management authorities to decide upon appropriate recovery solutions.

Flood susceptibility evaluation through deep learning optimizer ensembles and GIS techniques

It is difficult to predict and model with an accurate model the floods, that are one of the most destructive risks across the earth's surface. The main objective of this research is to show the prediction power of three ensemble algorithms with respect to flood susceptibility estimation. These algorithms are: Iterative Classifier Optimizer - Alternating Decision Tree - Frequency Ratio (ICO-ADT-FR), Iterative Classifier Optimizer - Deep Learning Neural Network - Frequency Ratio (ICO-DLNN-FR) and Iterative Classifier Optimizer - Multilayer Perceptron - Frequency Ratio (ICO-MLP-FR). The first stage of the manuscript consisted of the collection and processing of the geodatabase needed in the present study. The geodatabase comprises a number of 14 flood predictors and 132 known flood locations. The Correlation-based Feature Selection (CFS) method was used in order to assess the prediction capacity of the 14 predictors in terms of flood susceptibility estimation. The training and validation of the three ensemble models constitute the next stage of the scientific workflow. Several statistical metrics and ROC curve method were involved in the evaluation of the model's performance and accuracy. According to ROC curves all the models achieved high performances since their AUC had values above 0.89. ICO-DLNN-FR proved to be the most accurate model (AUC = 0.959). The outcomes of the study can be used to guide future flood risk management and sustainable land-use planning in the designated area.

Application of novel framework approach for prediction of nitrate concentration susceptibility in coastal multi-aquifers, Bangladesh

This study aims to construct a novel framework approach for predicting and mapping nitrate concentration susceptibility in the coastal multi-aquifers of Bangladesh by coupling the K-fold cross-validation method and novel ensemble learning algorithms, including Boosting, Bagging and Random Forest (RF). In total, 286 nitrate sampling sites were employed in the model work. The dataset was demarcated into a 75:25 ratio for model construction (75% 3-fold ≅ 214 sites) and (25% 1-fold ≅ 72 sites) for model validation using the 4-fold cross-validation schemes. A total of 14 groundwater causative factors including salinity, depth, pH, EC, As, HCO₃^-, F^-, Cl^-, SO₄^2-, PO₄^2-, Na⁺, K⁺, Mg²⁺, and Ca²⁺ were adopted for the construction of the proposed models. OneR relative importance model was employed to choose and rank critical factors for spatial nitrate modeling. The results showed that depth, pH and As are the most influential causative factors in the elevated nitrate concentration in groundwater. Based on the model assessment criteria such as receiver operating characteristic (ROC)'s AUC (area under curve), sensitivity, specificity, accuracy, precession, F score, and Kappa coefficient, the Boosting model outperforms others (r = 0.92, AUG ≥ 0.90) in mapping nitrate concentration susceptibility, followed by Bagging and RF models. The results of mapping nitrate concentration also demonstrated that the south-central and western regions had an elevated amount of nitrate content than other regions due to depth variation in the study area. During our sampling campaign, we observed hundreds of fish hatcheries operation, a fish landing center and aquaculture farms which are the reasons for overexploitation and excessive agrochemicals used in the study area. Thus, the dependability of ensemble learning modeling verifies the effectiveness and applicability of the proposed novel approach for decision-makers in groundwater pollution management at the local and regional levels.